commvita
Connected care platform
Artificial Intelligence

AI across commvita — useful, and governed

commvita uses AI where it saves time and never where it would take a clinical decision. There is one governed configuration (provider keys server-side only), AI at the point of care that suggests but never decides, executive decision-support with a mandatory trust envelope — and a governance spine (ATRS, DCB0129/0160, MHRA AIaMD, GDPR Art.22) over all of it. Non-SaMD throughout.

© 2026 Commvita Digital Health Solutions Ltd. All rights reserved.
Live vs demonstrated: Live — real, API-backed platform logic (wired end-to-end today) Demonstrated — representative control surface with seeded data / illustrative UI mock-up

Three ways commvita uses AI — one way it governs it

Configured once, used safely, evidenced always. No patient data leaves for a model without a server-side key and an audit entry; every AI surface is non-SaMD — it provides context and suggestions, a human makes the call.

One governed config

AI Integration Hub

The single, canonical AI configuration — Claude / Microsoft Copilot / OpenAI, one active provider. API keys are server-side environment only (never in the browser). The Ask commvita assistant and every AI feature call the same governed proxy.

Where in commvita /ai-integration-hub · API /ai/config · /ai/assistant/chat.

AI at the point of care

Ambient · scribe · coding

Ambient Clinical AI turns a consultation into a structured SOAP note with SNOMED auto-coding (audio never persisted); VoiceScribe, AI Consultation Assistant and NLP Document Coding assist documentation. Every code is accept/reject — the clinician decides.

Where in commvita /ambient-clinical-ai · /voice-scribe · /nlp-coding.

Executive decision-support

Transformation Advisor

The AI Transformation Advisor turns a strategic objective + constraints into options, prioritised interventions, KPIs and a roadmap — each with a mandatory trust envelope (assumptions, confidence, human-review areas, evidence) and a 3-way scenario comparison. Gated to executive + admin.

Where in commvita /ai-transformation-advisor · API /transformation/cases.
AI Integration Hub / Provider setup & governance admin
Keys are server-side only — supplied via environment (ANTHROPIC_API_KEY etc.), never entered in or returned to the browser.
ProviderModel / endpointKeyActive
Claude (Anthropic)claude-sonnet-4-6present● active
Microsoft Copilot (Azure OpenAI)gpt-4o · uksouthpresent
OpenAIgpt-4onot set
Governance: ATRS model card · DCB0129 hazard log · SHA-256 prompt audit (no prompt text) · GDPR Art.22 override log.
Representative UI — illustrative (config is live; secrets never shown)
1

Configure once, key never in the browser

Admins pick the active provider; the key comes from a server-side env var (mirrors the SMTP pattern). Test-connection checks key presence only. The assistant and all features read this one config.

/ai-integration-hub/ai/active-provider
2

Suggest, never decide

Point-of-care AI is non-SaMD: it drafts notes and proposes codes, and the clinician accepts, edits or rejects. Audio is transcript-only; the record is the clinician's.

Accept / reject per codeaudio not persisted
3

Evidence every use

The AI Governance module holds an ATRS model registry, DCB0129/0160 clinical-safety, NICE ESF tiers, MHRA AIaMD surveillance and a GDPR Art.22 override log; bias monitoring flags any disparity ratio outside 0.8–1.2 for DPO review.

/ai-governanceATRS · DCB0129 · Art.22

Governed, in evidence — registry, bias monitoring, impact assessments

The AI Governance module is where “is it safe and is it governed?” is answered with artefacts, not assurances: every tool carries a published model card (ATRS) and a NICE ESF risk tier; fairness is monitored quarterly against the Equality Act with a hard ±20% disparity threshold; and every deployment has an Algorithmic Impact Assessment covering DCB0129/0160, MHRA post-market surveillance and the devolved-nation gateways (Scotland PBPP, Wales HTW) plus the GDPR Art.22 right to human review.

AI Governance / Tool Registry (ATRS) clinical safety officer
Every AI tool has a published model card — purpose, tier, clinical-safety case and lawful basis, per the Algorithmic Transparency Recording Standard.
AI toolNICE ESF tierClinical safetyATRS
Ambient Clinical AITier C — InformDCB0129 ✓published
Deterioration ML (continuous RPM)Tier C — DriveDCB0129/0160 ✓published
Transformation AdvisorTier An/a (non-clinical)published
Ask commvita assistantTier An/a (non-clinical)published
Open model cardExport ATRS record
Representative UI — illustrative
AI Governance / Bias Monitoring data protection officer
1 disparity ratio outside 0.80–1.20 — PPV for one ethnic group is below tolerance. Flagged for mandatory DPO review (Equality Act 2010).action required
SubgroupMetricValueReferenceDisparityStatus
Age 75+Sensitivity0.860.900.96within
FemaleSensitivity0.880.900.98within
Ethnicity — BlackPPV0.710.900.79review
IMD decile 1Recall0.840.900.93within
Disparity ratio = subgroup metric ÷ reference. A value outside 0.80–1.20 trips a mandatory DPO review before the tool continues in that cohort.
Representative UI — illustrative
AI Governance / Algorithmic Impact Assessment admin
Tool · risk tier
Deterioration ML (continuous RPM) · NICE ESF Tier C — Drive
Lawful basis: UK GDPR Art.6(1)(e) public task · Art.9(2)(h) health · DPIA ref DPIA-2026-041
DCB0129 / DCB0160 clinical safety caseNamed Clinical Safety Officer · hazard log maintained
complete
MHRA AIaMD post-market surveillanceDrift detection & incident reporting (June 2025 regulations)
active
🏴󠁧󠁢󠁳󠁣󠁴󠁿 Scotland — Privacy Benefit Proportionality Purpose (PBPP)Panel review for Scottish deployment
complete
🏴󠁧󠁢󠁷󠁬󠁳󠁿 Wales — Health Technology Wales (HTW)Submitted for Welsh deployment
submitted
GDPR Art.22 — right to human reviewEvery AI-influenced decision is overridable; override log maintained
enforced
Sign off assessmentView override log
Representative UI — illustrative

Where it lives in commvita

CapabilityRouteModel / APIStatus
AI Integration Hub (single config)/ai-integration-hubAPI /ai/config · /ai/active-provider● Live
Ask commvita assistant (chat proxy)global overlayAPI /ai/assistant/chat · keyword fallback● Live
AI Transformation Advisor/ai-transformation-advisorAPI /transformation/cases · trust envelope● Live
Ambient Clinical AI (voice → SOAP)/ambient-clinical-aiSNOMED auto-code · hazard log☉ Demonstrated
VoiceScribe · AI Consult Assist/voice-scribe · /ai-consult-assistWeb Speech · suggestion panel☉ Demonstrated
NLP Document Coding/nlp-codingSNOMED · confidence · review queue☉ Demonstrated
AI Governance (ATRS · bias)/ai-governanceATRS · DCB0129/0160 · Art.22☉ Demonstrated
ML: continuous RPM · A&E demand · LOS/continuous-rpm · /ae-demand-forecastdecision-support (non-SaMD)☉ Demonstrated
Why this matters for a demo. The AI question a board will ask is “is it safe and is it governed?” commvita's answer is structural: one config with server-side keys, non-SaMD by design, and a governance spine (ATRS, DCB0129, MHRA AIaMD, GDPR Art.22, bias monitoring) that evidences every use. Humans stay in the loop.
© 2026 Commvita Digital Health Solutions Ltd. All rights reserved. ATRSDCB0129 / DCB0160NICE ESF Tiers A/B/CMHRA AIaMD Jun 2025 GDPR Art.22Equality Act 2010 (bias 0.8–1.2)Scotland PBPP · Wales HTWNon-SaMD